P3.C051: Harnessing pediatric primary care EHR data for more relevant and timely surveillance of obesity prevalence trends

Purpose/Objectives: Much of what is known about the burden of pediatric obesity is derived from national and/or state-level surveillance systems. The use of electronic health record (EHR) data is emerging as an alternative as it offers a low-cost, widely available means of obtaining more timely and locally relevant estimates. Our objective was to assess recent (pre- and post-COVID-19 pandemic) trends in obesity prevalence in a single pediatric primary care practice serving a primarily low-income minority population in Georgia and to compare locally obtained estimates to those obtained for the state through national and state surveillance systems.

Design/Methods: Data from children aged 2-19 years attending a pediatric primary care clinic for an annual well-child visit from 2016 to 2023 were used. BMI-for-age percentiles were calculated from patient height, weight, age, and sex information using CDC Growth Charts. Obesity (≥95th BMI-for-age percentile) prevalence was calculated overall and by sociodemographic characteristics for each year, and statistically significant linear trends were assessed for the following periods: pre-pandemic (2016-2019), pandemic (2019-2021), and post-pandemic (2021-2023) using R version 4.3.2. Results were compared against national estimates obtained from the National Health and Nutrition Examination Survey (NHANES) and estimates for the state of Georgia obtained from the National Survey of Children’s Health (NSCH) and WIC Participant and Program Characteristics. This quality improvement project was determined exempt from IRB review.

Results: On average, 6,102 patients were seen for well-child visits annually; 88.4% were Black, and 94.2% were Medicaid insured. In 2016, 17.9% of these children had obesity, and prevalence remained stable through 2019. Prevalence increased during the pandemic to a high of 24.0% in 2021. Prevalence has decreased significantly in the post-pandemic period, falling to 21.1% in 2023 (APC=-6.4%, p< 0.001) (Table). This trend over time was similar among all sociodemographic groups; however, obesity prevalence rates were highest among older children, Hispanic children, and children living in Spanish-speaking households. The obesity prevalence rates among children and adolescents 2–19 years attending well-child visits each year were similar (within the confidence intervals) to national estimates obtained in NHANES (data available through 2018). Notably, annual prevalence rates among 2–4-year-old children in the sample were lower than those obtained for same age children enrolled in Georgia’s WIC program (data available through 2020) and rates for the 10–17-year-olds in sample were significantly higher than the state average for same aged children as obtained using NSCH data (data available through 2022) (Table).

Conclusion/Discussion: Our results demonstrate the value of using EHR data to obtain timely and relevant estimates of obesity prevalence. These data can complement existing surveillance systems; EHR data can be used to identify and monitor trends in high-risk groups and to guide the development of and evaluate the effectiveness of obesity interventions.